CN115019498A

CN115019498A - Parking management method and system

Info

Publication number: CN115019498A
Application number: CN202210519670.9A
Authority: CN
Inventors: 陈烁; 高毅鹏; 黄凯明
Original assignee: Streamax Technology Co Ltd
Current assignee: Streamax Technology Co Ltd
Priority date: 2022-05-13
Filing date: 2022-05-13
Publication date: 2022-09-06

Abstract

The application discloses a parking management method and system, and belongs to the technical field of parking management. The method comprises the following steps: if a locking instruction of the vehicle is received, acquiring an environment image around the vehicle, acquired by an image acquisition device configured for the vehicle, determining structural information of a visual positioning identifier in the environment image through the identifier identification model by taking the environment image as input of the identifier identification model, then determining whether the vehicle is located in a designated parking area according to the structural information, and if the vehicle is determined to be located in the designated parking area, locking the vehicle according to the locking instruction. Wherein the structured information indicates an identification characteristic of the visual positioning identification. Therefore, the visual positioning mark is preset in the designated parking area, and whether the vehicle is located in the designated parking area is determined according to the visual positioning mark in the environment image around the vehicle, so that the problems of poor positioning accuracy, mistaken locking of a non-motor vehicle or failure in locking and the like caused by drift of a GPS signal can be avoided.

Description

Parking management method and system

Technical Field

The present application relates to the field of parking management technologies, and in particular, to a parking management method and system.

Background

With the development of 'shared economy', industries such as express delivery and take-out and the like rise, the problem of random parking and random placement of non-motor vehicles such as bicycles and electric vehicles draws more and more attention. In order to maintain the city appearance of a city, a reliable means is needed to standardize parking of non-motor vehicles.

In the prior art, parking management of non-motor vehicles is generally performed in conjunction with a GPS (Global Positioning System). For example, for the shared bicycle, after a locking instruction of a user is received, a GPS signal of the shared bicycle may be acquired first, whether the shared bicycle is located in an appointed parking area is determined through the GPS signal, and if the shared bicycle is located in the appointed parking area, the shared bicycle is locked according to the locking instruction, so as to achieve the purpose of managing the standard parking of the shared bicycle.

However, in the above-described method of managing parking of a non-motor vehicle in combination with GPS, there is a problem that the GPS signal is shifted in a complicated environment area such as a dense building, and the positioning accuracy is poor, and therefore, the non-motor vehicle may be locked by mistake or may fail to lock the vehicle.

Disclosure of Invention

The application provides a parking management method and system, which can avoid the problems that in a complex environment area, due to the fact that GPS signals drift, the positioning accuracy is poor, and a non-motor vehicle is locked by mistake or the locking fails. The technical scheme is as follows:

in a first aspect, a parking management method is provided, the method comprising:

if a locking instruction of a vehicle is received, acquiring an environment image around the vehicle, which is acquired by an image acquisition device configured for the vehicle;

taking the environment image as an input of an identification recognition model, and determining structural information of a visual positioning identification in the environment image through the identification recognition model, wherein the visual positioning identification is arranged in a specified parking area, and the structural information is used for indicating an identification feature of the visual positioning identification;

determining whether the vehicle is located within the designated parking area based on the structured information;

and if the vehicle is determined to be located in the designated parking area, locking the vehicle according to the locking instruction.

As one example, the visual location identifier includes at least one location identifier, and the structured information includes an identifier category for each of the at least one location identifier, and a detection box and a confidence of the detection box in the environmental image;

the determining whether the vehicle is located within the designated parking area according to the structured information includes:

determining whether the vehicle is located in the designated parking area according to the identification category of each positioning identification in the at least one positioning identification and the confidence degree of the detection frame

As an example, the determining whether the vehicle is located in the designated parking area according to the identification category of each of the at least one positioning identification and the confidence of the detection box includes:

determining the weight corresponding to each positioning identifier according to the identifier category of each positioning identifier in the at least one positioning identifier;

according to the confidence degree of the detection frame of each positioning identifier in the at least one positioning identifier and the weight corresponding to each positioning identifier, performing weighted summation on the confidence degrees of the detection frames of the at least one positioning identifier to obtain the confidence degree that the vehicle is located in the designated parking area:

if the confidence degree that the vehicle is located in the designated parking area is greater than a confidence degree threshold value, determining that the vehicle is located in the designated parking area;

if the confidence degree that the vehicle is positioned in the designated parking area is less than or equal to the confidence degree threshold value, determining that the vehicle is not positioned in the designated parking area

As one example, before the locking the vehicle according to the locking instruction, the method further comprises:

determining whether the parking direction of the vehicle meets the direction requirement and whether the parking posture of the vehicle meets the posture requirement according to the structured information;

if the vehicle is determined to be located in the designated parking area, the parking direction meets the direction requirement, and the parking posture meets the posture requirement, the step of locking the vehicle according to the locking instruction is executed

As one example, the visual location identifier includes at least one location identifier, and the structured information includes an identifier category for each of the at least one location identifier, and pixel locations, detection boxes, and confidence levels for the detection boxes in the environmental image;

the determining whether the parking direction of the vehicle meets the direction requirement and the parking posture of the vehicle meets the posture requirement according to the structured information includes:

determining whether the parking direction of the vehicle meets the direction requirement or not according to the identification category of each positioning identification in the at least one positioning identification and the pixel position in the environment image;

determining whether the parking gesture of the vehicle meets the gesture requirement or not according to the detection frame of each positioning identifier in the at least one positioning identifier in the environment image and the confidence of the detection frame

As an example, the determining whether the parking direction of the vehicle meets the direction requirement according to the identifier category and the pixel position in the environment image of each of the at least one location identifier includes:

determining the position relation between the at least one positioning identifier according to the identifier category of each positioning identifier in the at least one positioning identifier and the pixel position in the environment image;

if the position relation meets a first preset condition, determining that the parking direction of the vehicle meets the direction requirement;

and if the position relation does not meet the first preset condition, determining that the parking direction of the vehicle does not meet the direction requirement.

As an example, the determining whether the parking posture of the vehicle meets the posture requirement according to the detection frame and the confidence of the detection frame of each of the at least one positioning identifier in the environment image includes:

determining a positioning identifier with the maximum confidence coefficient of the detection frame from the at least one positioning identifier as a target positioning identifier;

determining the parking posture of the vehicle according to the width and the height of the detection frame of the target positioning identifier;

if the parking posture meets a second preset condition, determining that the parking posture of the vehicle meets the posture requirement;

and if the parking posture does not accord with the second preset condition, determining that the parking posture of the vehicle does not accord with the posture requirement.

As one example, the identity recognition model includes an identity detection module and an identity classification module, the visual location identity includes at least one location identity, and the structured information includes an identity category of each of the at least one location identity, and pixel locations, detection boxes, and confidence levels of the detection boxes in the environmental image;

the taking the environment image as an input of an identification recognition model, and determining structural information of a visual positioning identification in the environment image through the identification recognition model comprises:

taking the environment image as the input of the identification detection module, and performing feature extraction on the environment image through the identification detection module to obtain the pixel position of each positioning identification in the at least one positioning identification, a detection frame and the confidence of the detection frame;

and taking the pixel position of each positioning identifier, the detection frame and the confidence coefficient of the detection frame in the at least one positioning identifier as the input of the identifier classification module, determining the identifier attribute of each positioning identifier through the identifier classification module, and determining the identifier category of each positioning identifier according to the identifier attribute of each positioning identifier, wherein the identifier attribute comprises one or more of color, shape, width and height and texture.

As an example, before the environmental image is used as an input of an identification recognition model, and the structural information of the visual positioning identification in the environmental image is determined by the identification recognition model, the method further comprises:

acquiring a sample environment image around a sample vehicle;

labeling the sample environment image to obtain sample structural information of the visual positioning identifier in the sample environment image, wherein the sample structural information is used for indicating a sample identifier feature of each positioning identifier in at least one positioning identifier;

and training a to-be-trained identification model according to the sample environment image and the sample structural information to obtain the identification model, wherein the identification model is used for identifying the structural information of the visual positioning identification in any environment image.

As an example, the training a to-be-trained identifier recognition model according to the sample environment image and the sample structural information to obtain the identifier recognition model includes:

inputting the sample environment image into the identification recognition model to be trained, and determining the prediction structural information of the visual positioning identification in the sample environment image through the identification recognition model to be trained;

determining the recognition error of the identification model to be trained according to the sample structural information and the prediction structural information;

and adjusting the model parameters of the identification model to be trained according to the identification error, and taking the identification model to be trained after the model parameters are adjusted as the identification model.

In a second aspect, a parking management system is provided, which includes an image acquisition device and a main control unit, wherein the image acquisition device is fixedly connected with a vehicle;

the image acquisition device is used for acquiring an environment image around the vehicle;

the main control unit is used for acquiring an environment image around the vehicle, which is acquired by an image acquisition device configured by the vehicle, if a locking instruction of the vehicle is received;

the main control unit is further configured to use the environment image as an input of an identifier recognition model, and determine structural information of a visual positioning identifier in the environment image through the identifier recognition model, where the visual positioning identifier is arranged in a specified parking area, and the structural information is used to indicate an identifier feature of the visual positioning identifier;

the main control unit is further used for determining whether the vehicle is located in the designated parking area according to the structured information;

the main control unit is further configured to lock the vehicle according to the locking instruction if it is determined that the vehicle is located in the designated parking area.

As an example, the collection angle of the image collection device is within a preset angle range, and the preset angle range is used for controlling the environment image collected by the image collection device to include the ground environment in front of the vehicle.

The technical scheme provided by the embodiment of the application has the following beneficial effects:

in the embodiment of the application, if a locking instruction of a vehicle is received, an environment image around the vehicle, which is acquired by an image acquisition device configured for the vehicle, is acquired, then the environment image is used as an input of an identification recognition model, structural information of a visual positioning identification in the environment image is determined through the identification recognition model, then whether the vehicle is located in a specified parking area is determined according to the structural information, and if the vehicle is determined to be located in the specified parking area, the vehicle is locked according to the locking instruction. Wherein the structured information is used to indicate an identification feature of the visual positioning identification. Therefore, the visual positioning identifier is preset in the designated parking area, and whether the vehicle is located in the designated parking area is determined according to the visual positioning identifier in the acquired environment image around the vehicle, so that the problems that the positioning accuracy is poor, the non-motor vehicle is locked by mistake or the vehicle locking fails and the like due to the drift of the GPS signal in the complicated environment area can be avoided.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a system diagram of a parking management system provided in an embodiment of the present application;

fig. 2 is a flowchart of a parking management method according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.

It should be understood that reference to "a plurality" in this application means two or more. In the description of the present application, "/" means "or" unless otherwise stated, for example, a/B may mean a or B; "and/or" herein is only an association relationship describing an associated object, and means that there may be three relationships, for example, a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, for the convenience of clearly describing the technical solutions of the present application, the words "first", "second", and the like are used to distinguish the same items or similar items having substantially the same functions and actions. Those skilled in the art will appreciate that the terms "first," "second," etc. do not denote any order or quantity, nor do the terms "first," "second," etc. denote any order or importance.

Before explaining the embodiments of the present application in detail, an application scenario of the embodiments of the present application will be described.

The non-motor vehicles such as bicycles, electric vehicles and the like can realize the function of unlocking or locking the bicycle anytime and anywhere, and the problem of the last kilometer is solved, so that the non-motor vehicles become one of important tools in people's travel and life. But the appearance of non-motor vehicles also brings much inconvenience to city management and brings great influence to city appearance of cities. For example, the user can park and overturn the non-motor vehicle at will, which not only increases the workload of the maintenance and placement personnel, but also brings great pressure to the city management. To maintain the appearance of a city and reduce the burden of city management, a reliable means is needed to regulate parking of non-motor vehicles.

In the prior art, a GPS (Global Positioning System) technology is usually combined to perform parking management on a non-motor vehicle, so as to solve the problem of random parking and random parking of the non-motor vehicle. For example, for a shared bicycle, parking management of the shared bicycle in the prior art is usually based on a GPS signal on the vehicle, and determines whether the vehicle enters a designated parking area through the GPS signal, and locks the shared bicycle according to a vehicle locking instruction if it is determined that the vehicle enters the designated parking area.

However, in the above-described method of managing parking of a non-motor vehicle in combination with GPS, there is a problem that the GPS signal is shifted in a complicated environment area such as a dense building, and the positioning accuracy is poor, and therefore, the non-motor vehicle may be locked by mistake or may fail to lock the vehicle. For example, the shared vehicle is erroneously determined to be located in the designated parking area due to poor GPS positioning accuracy, thereby resulting in an erroneous vehicle lock.

Based on this, the embodiment of the application provides a parking management method, and by presetting a visual positioning identifier in a designated area and determining whether a vehicle is located in the designated parking area according to the visual positioning identifier in an acquired environment image around the vehicle, the problems that the positioning accuracy is poor, a non-motor vehicle is locked by mistake or the vehicle locking fails and the like due to drift of a GPS signal in a complex environment area can be avoided. In addition, whether the parking direction and the parking gesture of the vehicle meet the specifications or not can be determined according to the visual positioning identification in the environment image, so that the non-motor vehicle is further subjected to parking management, the influence of the parking direction and the parking gesture of the vehicle on city appearance and city appearance of a city is reduced, and the waste of parking space resources is reduced.

A parking management system provided in an embodiment of the present application is explained in detail below.

Referring to fig. 1, fig. 1 is a system schematic diagram of a parking management system according to an embodiment of the present disclosure. As shown in fig. 1, the parking management system includes an image capturing device 101 and a main control unit 102, and the parking management method provided in the embodiment of the present application may be applied to the parking management system in fig. 1.

The vehicle can be a bicycle, an electric vehicle and other non-motor vehicles, and the function of unlocking or locking the vehicle at any time and any place can be realized. For example, as shown in fig. 1, the vehicle is a bicycle. The vehicle in fig. 1 is only an example, and is not limited to the vehicle.

The image capturing device 101 is fixedly connected to the vehicle and is configured to capture an environmental image around the vehicle.

The ambient image may be in the form of a photograph. In addition, the image capturing module 101 may also capture an environment around the vehicle in a video form, that is, the image capturing module 101 may also capture an environment video around the vehicle, which is not limited in this embodiment of the application.

The image capturing device 101 may be a camera with a shooting function or a camera with a monitoring function, and the image capturing device 101 has a capturing angle and a capturing range (area).

For example, an included angle between the image capturing device 101 and the vehicle body in the vertical direction is a capturing angle, the capturing angle of the image capturing device 101 is within a preset angle range, and the preset angle range is used for controlling an environment image captured by the image capturing device 101 to include a ground environment in front of the vehicle.

As an example, the image capturing device 101 may be directly fixedly connected to the vehicle, or may be fixedly connected to the vehicle through a connecting member, for example, by welding.

As one example, the image pickup device 101 does not rotate with the rotation of the head of the vehicle, and the posture of the image pickup device 101 is kept in agreement with the posture of the body of the vehicle.

The main control unit 102 is configured to acquire an environment image around the vehicle, which is acquired by the image acquisition device 101 configured in the vehicle, if a locking instruction of the vehicle is received.

The main control unit 102 is further configured to use the environment image as an input of an identifier recognition model, and determine structural information of a visual positioning identifier in the environment image through the identifier recognition model, where the visual positioning identifier is disposed in the designated parking area, and the structural information is used to indicate an identifier feature of the visual positioning identifier.

The master control unit 102 is also configured to determine whether the vehicle is located within the designated parking area based on the structured information.

The main control unit 102 is further configured to lock the vehicle according to the locking instruction if it is determined that the vehicle is located in the designated parking area.

The master control unit 102 may be a device with processing capability, for example, the master control unit 102 may be a computer device or a processor with processing capability.

As an example, the master control unit 102 may be fixedly connected to the vehicle directly or may be connected to the vehicle through a connector, in which case the master control unit 102 may be a processor with processing capability. Or, the main control unit 102 is not connected to the vehicle, and the main control unit 102 may remotely obtain the environment image around the vehicle collected by the image collecting module 101, and determine whether the vehicle is located in the designated parking area according to the environment image, in which case, the main control unit 102 may be a computer device, and the computer device may be a terminal or a server.

For example, as shown in fig. 1, the main control unit 102 is fixedly connected to the vehicle through a connector.

The main control unit 102 and the image capturing module 101 may be connected in a wired or wireless manner. For example, the main control unit 102 is connected to a vehicle, and the main control unit 102 may be connected to the image capturing module 101 in a wired manner.

In addition, a visual positioning mark can be arranged in the designated parking area in advance, namely, the visual positioning mark can be arranged in a special area or a special position in the designated parking area. The visual positioning mark can be pasted on the designated parking area or printed on the ground of the designated parking area, and the setting form of the visual positioning mark is not limited in the embodiment of the application.

The visual positioning mark is a graph for carrying out visual positioning. The visual positioning mark has the characteristics of easy recognition, easy expansion and the like. For example, as shown in fig. 1, visual positioning indicators may be printed on the floor of the edge of the designated parking area, including a first positioning indicator 1031 and a second positioning indicator 1032. First location sign 1031 and second location sign 1032 are rectangular printing signs, but first location sign 1031 and second location sign 1032 have different width and height, and first location sign 1031 and second location sign 1032 have different texture characteristics.

In addition, the first and

second location indicators

1032, 1032 may have different colors. For example, the first positioning indicator 1031 is yellow, and the second positioning indicator 1032 is white.

Since the visual positioning identifier is located in the designated parking area, if the vehicle is located in the designated parking area or if the vehicle is located in a vicinity of the designated parking area, the image capturing device 101 may include the visual positioning identifier in the acquired environment image, and the main control unit 102 may determine whether the vehicle is located in the designated parking area according to the visual positioning identifier in the acquired environment image around the vehicle, for example, determine whether the vehicle is located in the designated parking area according to the structural information of the visual positioning identifier. Therefore, the problems that in a complex environment area, due to the fact that GPS signals drift, positioning accuracy is poor, and a non-motor vehicle is locked by mistake or fails in locking can be solved.

Further, as shown in fig. 1, the parking management system may further include at least one of a locking device 104, a communication device 105, and an alarm device 106.

The locking device 104 is used to unlock or lock the vehicle.

As one example, the locking device 104 may include a locking module. The locking module is used for locking the vehicle according to the locking instruction or unlocking the vehicle according to the unlocking instruction. For example, the locking module may be a mechanical device for achieving mechanical unlocking or mechanical locking of the vehicle.

As one example, the locking device 104 may also include a detection module. The detection module is configured to detect a locking operation of a user, and send a locking request to the main control unit 102 if the locking operation is detected. The main control unit 102 receives the locking request sent by the detection module, generates a locking instruction according to the locking request, and sends the locking instruction to the locking module.

The locking device 104 may be fixed to the vehicle, and the main control unit 102 and the locking device 104 may be connected in a wired manner or a wireless manner.

As an example, the locking device 104 may be fixedly connected to the vehicle directly or via a connecting member, for example, by welding.

Wherein the communication means 105 is used for communicating with systems or devices outside the parking management system.

For example, the communication device 105 may receive a lock instruction sent outside the system. Upon receiving a lock command transmitted from the outside of the system, the communication device 105 transmits the lock command to the master control unit 102. The master control unit 102 receives the lock instruction and sends the lock instruction to the locking device 104. Thereafter, the lock device 104 locks the vehicle according to the lock instruction.

The communication device 105 may be fixed to the vehicle, the main control unit 102 and the communication device 105 may be connected in a wired manner or a wireless manner, and the communication device 105 and a device other than the parking management system may be connected in a wired manner or a wireless manner.

As an example, the communication device 105 may be directly fixedly connected to the vehicle, or may be fixedly connected to the vehicle through a connector, for example, by welding.

The alarm device 106 is used for sending alarm information.

For example, the alarm device 106 may be a speaker, and the alarm device 106 sends an alarm message through the speaker.

Next, a parking management method provided in an embodiment of the present application will be described.

Referring to fig. 2, fig. 2 is a flowchart of a parking management method according to an embodiment of the present disclosure, where the method may be applied to the parking management system shown in fig. 1, and the parking management system may include an image capturing device and a main control unit, where the image capturing device is fixedly connected to a vehicle. As shown in fig. 2, the method comprises the steps of:

step 201, if receiving a locking instruction of a vehicle, a main control unit sends an image acquisition instruction to an image acquisition device configured for the vehicle.

The locking instruction is used for indicating the main control unit to lock the vehicle, and the acquired image instruction is used for indicating the image acquisition device to acquire the environment image around the vehicle.

Wherein, the locking instruction can be triggered by a user through locking operation or automatically triggered by the main control unit.

As an example, as shown in fig. 1, the parking management system further includes a communication device. And if the communication device receives a locking instruction sent from the outside of the system, the communication device sends the locking instruction to the main control unit. And then, the main control unit receives the locking instruction sent by the communication device.

Wherein, the locking instruction sent outside the system is triggered by the locking operation of the user. For example, the communication device communicates with a cell phone outside the parking management system. The mobile phone can execute locking operation when a user needs to lock the vehicle, and the mobile phone sends a locking instruction to the communication device if detecting the locking operation. The communication device receives a locking instruction sent by the mobile phone and sends the locking instruction to the main control unit. That is, when the user needs to lock the vehicle, the main control unit obtains the lock instruction.

The locking operation may be executed by a user, and the operation type of the locking operation may be a click operation, a press operation, a language operation, or a gesture operation, which is not limited in this embodiment of the application.

The main control unit and the communication device can be connected in a wired or wireless mode, and the communication device and equipment outside the parking management system can be connected in a wired or wireless mode.

As one example, as shown in fig. 1, the parking management system further includes a locking device that transmits a locking request to the main control unit if a locking operation performed by a user is detected. The main control unit receives a locking request sent by the locking device and generates a locking instruction according to the locking request. Wherein the locking instruction can be automatically triggered by the main control unit according to the locking request.

For example, the locking device may include a detection module, and the user may perform a locking operation through the detection module in a case where the user needs to lock the vehicle. And then, if the detection module detects the locking operation executed by the user, the detection module sends a locking request to the main control unit. The main control unit receives a locking request sent by the locking device and generates a locking instruction according to the locking request. That is, when the user needs to lock the vehicle, the main control unit obtains the locking instruction.

For example, the detection module is a lock button, the user may perform a locking operation by pressing the lock button, and the lock button may detect the locking operation and send a lock request to the main control unit.

The locking device can send a locking request to the main control unit in a wired or wireless mode.

The main control unit can be connected with the image acquisition device in a wired or wireless mode, and can send an image acquisition instruction to the image acquisition device in a wired or wireless mode.

Step 202, the image acquisition device receives an image acquisition instruction sent by the main control unit and acquires an environmental image around the vehicle.

The image acquisition device acquires the environment around the vehicle when the user needs to lock the vehicle. For example, when the user needs to lock the vehicle, the main control unit obtains the lock instruction and sends an image obtaining instruction to the image acquisition device. And if the image acquisition device receives the image acquisition command, acquiring an environmental image around the vehicle.

The image acquisition device is fixedly connected with the vehicle, the image acquisition device is provided with an acquisition angle and a visual field range (visual field area), the acquisition angle is within a preset angle range, and the preset angle range is used for controlling an environment image acquired by the image acquisition device to comprise a ground environment in front of the vehicle. For example, the image capturing device captures an environmental image in a field of view according to the capturing angle.

As an example, the connection position of the image capturing device to the vehicle affects the capturing angle. For example, if the image capturing device is connected to a position on the vehicle closer to the ground, the capturing angle is larger, and if the image capturing device is connected to a position on the vehicle farther from the ground, the capturing angle is smaller, so as to ensure that the image capturing device captures the ground environment in front of the vehicle.

In addition, a visual positioning mark can be arranged in the designated parking area in advance, the visual positioning mark can be pasted on the designated parking area or printed on the ground of the designated parking area, and the visual positioning mark is not limited in the embodiment of the application.

Because the visual positioning mark is positioned in the designated parking area, if the vehicle is positioned in the designated parking area or if the vehicle is positioned in the vicinity of the designated parking area, the visual positioning mark may be included in the environment image acquired by the image acquisition device.

Step 203, the image acquisition device sends an environment image to the main control unit.

The image acquisition device can send the environment image to the main control unit in a wired mode or a wireless mode.

And step 204, the main control unit receives the environment image sent by the image acquisition device.

Since the visual positioning identifier is located in the designated parking area, and if the vehicle is located in the designated parking area, or if the vehicle is located in a vicinity area of the designated parking area, the environment image acquired by the image acquisition device may include the visual positioning identifier, the main control unit may determine whether the vehicle is located in the designated parking area according to the visual positioning identifier in the acquired environment image around the vehicle, for example, determine whether the vehicle is located in the designated parking area according to the structural information of the visual positioning identifier. Therefore, the method can avoid determining whether the vehicle is positioned in the designated parking area by combining the GPS, namely avoid parking management of the non-motor vehicle by combining the GPS, thereby avoiding the problems of poor positioning accuracy, mistaken locking of the non-motor vehicle or failure in locking and the like caused by drift of GPS signals in a complicated environment area.

In addition, if the main control unit receives a locking instruction of the vehicle, the step 201 to the step 204 may be implemented to acquire an environment image around the vehicle, which is acquired by the image acquisition device configured in the vehicle, or may be implemented in other manners.

For example, the environment image acquired by the main control unit after receiving the locking instruction of the vehicle may be referred to as a target environment image. The image acquisition device can acquire the environment image around the vehicle every preset time period after the vehicle is unlocked and send the environment image to the main control unit. The main control unit receives the environment image sent by the image acquisition device and stores the environment image. And if the main control unit receives the locking instruction, taking the environment image which is sent by the image acquisition device for the first time after the locking instruction is received as the target environment image.

In step 205, the main control unit uses the environment image as an input of the identification recognition model, and determines the structural information of the visual positioning identification in the environment image through the identification recognition model.

The visual positioning mark is arranged in the appointed parking area, and the structured information is used for indicating the mark characteristics of the visual positioning mark. The identification recognition model is used for recognizing the structural information of the visual positioning identification in any environment image.

Because the visual positioning identifier is set in the designated parking area in advance, if the vehicle is located in the designated parking area or if the vehicle is located in the vicinity of the designated parking area, the environmental image acquired by the image acquisition device may include the visual positioning identifier, that is, the environmental image acquired by the main control unit may include the visual positioning identifier.

The visual positioning mark is arranged in the designated parking area, namely the visual positioning mark is arranged in a special area or a special position in the designated parking area. For example, visual positioning marks are arranged on the periphery of the parking area. The visual positioning mark can be pasted on the designated parking area or printed on the ground of the designated parking area, and the visual positioning mark is not limited in the embodiment of the application.

The visual positioning identifier may include at least one positioning identifier, and each of the at least one positioning identifier has an identifier category. The structured information includes an identification category for each of the at least one location identification, and a detection box and a confidence level for the detection box in the environmental image.

For example, the visual positioning mark is printed on the ground at the edge of the designated parking area, and the visual positioning mark comprises a first positioning mark and a second positioning mark. The first positioning mark is a yellow rectangular printed mark having a first width and a first height, and the second positioning mark is a white rectangular printed mark having a second width and a second height.

The identification category is used for uniquely identifying the positioning identifier, for example, the identification category may be a color of the positioning identifier, for example, the identification category of the first positioning identifier is yellow, and the identification category of the second positioning identifier is white.

The detection frame refers to a circumscribed rectangle of the positioning identifier in the environment image, and the confidence coefficient of the detection frame refers to the probability that the detection frame is the detection frame of the positioning identifier. For example, the confidence level of the detection frame of the first positioning identifier in the environment image refers to the probability that the detection frame is the detection frame of the first positioning identifier.

In addition, the structured information may further include a pixel position of each of the at least one localization marker in the environmental image.

For example, the pixel position of the first positioning identifier in the environment image may refer to a plurality of pixel positions of the first positioning identifier in the environment image, which are determined by the main control unit through the identifier recognition model, or refer to a central pixel position of the first positioning identifier in the environment image, or refer to four vertex pixel positions of the detection frame of the first positioning identifier in the environment image.

Furthermore, each of the at least one location indicator has an identification attribute, which may include one or more of color, shape, width and height, and texture. For example, the main control unit may determine the identifier category according to the identifier attribute of the positioning identifier.

As one example, the visual positioning indicia is printed on the floor of the designated parking area, the visual positioning indicia includes a first positioning indicia and a second positioning indicia, the indicia attributes of the first positioning indicia include yellow, rectangular, a first width and a first height, the indicia attributes of the second positioning indicia include white, rectangular, a second width and a second height, the first width is equal to the second width, and the first height is less than the second height. The main control unit may identify an identifier attribute of each of the at least one location identifier in the environmental image, and determine an identifier category of each of the at least one location identifier in the environmental image according to the identifier attribute.

As an example, at least one positioning identifier has a certain positional relationship therebetween. For example, the position relationship between at least one positioning mark conforms to a first preset condition, the visual positioning mark includes a first positioning mark and a second positioning mark, and the first preset condition is that the printing position area of the first positioning mark is closer to the boundary of the designated parking area, that is, the first preset condition is that the pixel position of the first positioning mark in the environmental image is smaller than the pixel position of the second positioning mark. And the origin of coordinates of the pixel coordinate system corresponding to the environment image is positioned at the upper left corner of the environment image.

The identification recognition model can comprise an identification detection module and an identification classification module, the visual positioning identification comprises at least one positioning identification, and the structured information comprises an identification category of each positioning identification in the at least one positioning identification, and pixel positions, detection frames and confidence degrees of the detection frames in the environment image. The main control unit takes the environment image as the input of the identification recognition model, and the determination of the structural information of the visual positioning identification in the environment image through the identification recognition model can be realized through the following steps:

step 2051, the main control unit takes the environment image as an input of the identifier detection module, and performs feature extraction on the environment image through the identifier detection module to obtain a pixel position of each location identifier in the at least one location identifier, a detection frame and a confidence of the detection frame.

The identification detection module is configured to detect identification features of the visual positioning identifications in any environment image, such as pixel positions of each positioning identification in at least one positioning identification in any environment image, a detection frame, and a confidence of the detection frame.

The identification detection module may perform feature extraction on the environment image through corner detection, spot detection, a color histogram, a gray level co-occurrence matrix, a shape invariant moment method, or the like, for example, perform feature extraction on a color feature, a texture feature, a shape feature, or a local feature of the environment image.

After the feature extraction is performed on the environmental image by the identifier detection module, the similarity between features can be calculated, and the pixel position of each positioning identifier in at least one positioning identifier in the environmental image, the confidence of the detection frame and the confidence of the detection frame are determined according to the similarity.

Step 2052, the main control unit takes the pixel position of each positioning identifier in the at least one positioning identifier, the detection frame and the confidence coefficient of the detection frame as the input of the identifier classification module, determines the identifier attribute of each positioning identifier through the identifier classification module, and determines the identifier category of each positioning identifier according to the identifier attribute of each positioning identifier.

Wherein the identification attribute comprises one or more of color, shape, width and height, and texture.

The identification classification module is used for determining the identification category of any positioning identification in the environment image. For example, the identifier classification module determines an attribute of any positioning identifier in the environment image, and determines an identifier category according to the identifier attribute.

As an example, the identification recognition model may be obtained by training the sample environment image in advance, and then a model training process of the identification recognition model is described. For example, the model training process for identifying recognition models may include the following steps:

step 1) a main control unit obtains a sample environment image around a sample vehicle.

The system comprises a main control unit, a sample vehicle, a sample image acquisition device and a sample image acquisition device, wherein the sample vehicle is a non-motor vehicle, the sample vehicle is provided with the sample image acquisition device, and the main control unit acquires a sample environment image acquired by the sample image acquisition device arranged on the sample vehicle.

The main control unit can acquire a plurality of sample environment images of each of a plurality of sample vehicles.

The system comprises a main control unit, a sample vehicle, a designated parking area and a visual positioning identifier, wherein the designated parking area of the sample vehicle is also preset with the visual positioning identifier, so that if the sample vehicle is located in the designated parking area or if the sample vehicle is located in the area close to the designated parking area, the sample environment image acquired by the main control unit may include the visual positioning identifier.

And step 2) the main control unit labels the sample environment image to obtain sample structural information of the visual positioning identifier in the sample environment image.

Wherein the sample structural information is used for indicating the sample identification characteristics of each positioning identification in the at least one positioning identification.

For example, the sample structured information may include a sample identification category for each of the at least one location identification, and a sample pixel location in the sample environment image, a sample detection box, and a sample confidence for the sample detection box.

For example, each sample image in the multiple sample images is manually labeled, and the sample structural information in each sample environment image is labeled.

As an example, the identity attribute of each of the at least one location identity may also be labeled.

And 3) training the identification recognition model to be trained by the main control unit according to the sample environment image and the sample structural information to obtain the identification recognition model.

The identification recognition model is used for recognizing the structural information of the visual positioning identification in any environment image.

For example, the main control unit may input the sample environment image into the identification recognition model to be trained, and determine the predicted structural information of the visual positioning identifier in the sample environment image through the identification recognition model to be trained. And determining the recognition error of the identification model to be trained through the artificially marked sample structural information and the predicted structural information. And then, adjusting the model parameters of the identification model to be trained according to the identification errors, and taking the identification model to be trained after the model parameters are adjusted as the identification model.

As an example, a recognition error loss function of the sample structured information and the prediction structured information may be constructed, and the model parameters of the identification model to be trained may be adjusted according to the recognition error loss function. For example, the model parameters are optimized using a gradient descent algorithm for the identification error loss function.

As an example, the identification recognition model to be trained may be a multitasking network based on a neural network, such as an FCOS (full volumetric One-Stage Object Detection) based multitasking network. An FCOS-based multitasking network may include a Backbone module, a NECK module, and a Head module. The backhaul module performs convolution pooling operation for multiple times, is a feature extraction layer, inputs a sample environment image and outputs a plurality of different feature layers. The Neck module is a feature fusion layer, the input is a plurality of different feature layers, and the output is a feature layer fusing different depths. The Head module is an output layer of the identification model to be trained, the input of the Head module is a feature layer with different depths, and the output of the Head module is a plurality of output heads Head. Wherein the plurality of output heads Head indicate structured information.

For example, the structured information includes an identification category of each of the at least one location identification, and confidence levels of the pixel position, the detection frame, and the detection frame in the environment image, the output of the Head module includes a first Head, a second Head, and a third Head, the first Head indicates the identification category, the second Head indicates the pixel position, and the third Head indicates the confidence levels of the detection frame and the detection frame.

It should be noted that the above-mentioned identification recognition model to be trained is only an example, and is not a limitation of the identification recognition model to be trained. The identification model to be trained can also be other neural network models, and the identification model to be trained and the identification model are not limited in the embodiment of the application.

As an example, if the number of times of adjusting the model parameters is determined to be equal to the preset number of times, or if the recognition error is determined to be less than or equal to the error threshold, the main control unit stops adjusting the model parameters, and takes the identification model to be trained of the model parameters adjusted last time as the identification recognition model.

The steps 1) to 3) of the model training process for identifying the recognition model may be executed by a computer device, and the computer device may be a terminal or a server. The main control unit can store the trained identification recognition model, so that the structural information of the visual positioning identification in the environment image is determined through the identification recognition model.

In step 206, the master control unit determines whether the vehicle is located within the designated parking area based on the structured information.

The main control unit may determine whether the vehicle is located in the designated parking area according to the visual positioning identifier in the acquired environmental image around the vehicle, for example, determine whether the vehicle is located in the designated parking area according to the structural information of the visual positioning identifier. Therefore, the situation that whether the vehicle is located in the designated parking area or not can be determined by combining the GPS, namely, the situation that parking management is carried out on the non-motor vehicle by combining the GPS is avoided, and the problems that the positioning accuracy is poor, the non-motor vehicle is locked by mistake or the vehicle locking fails and the like due to the fact that the GPS signal drifts in a complex environment area are avoided.

The visual positioning identifier comprises at least one positioning identifier, the structured information comprises an identifier category of each positioning identifier in the at least one positioning identifier, and a detection frame and a confidence degree of the detection frame in the environment image, and the main control unit can determine whether the vehicle is located in the designated parking area according to the identifier category of each positioning identifier in the at least one positioning identifier and the confidence degree of the detection frame.

For example, the main control unit may determine the weight corresponding to each positioning identifier according to the identifier category of each positioning identifier in the at least one positioning identifier. And then carrying out weighted summation on the confidence degrees of the detection frames of the at least one positioning identifier according to the confidence degree of the detection frame of each positioning identifier in the at least one positioning identifier and the weight corresponding to each positioning identifier to obtain the confidence degree of the vehicle in the designated parking area. And then, if the confidence coefficient that the vehicle is positioned in the designated parking area is greater than the confidence coefficient threshold value, determining that the vehicle is positioned in the designated parking area, and if the confidence coefficient that the vehicle is positioned in the designated parking area is less than or equal to the confidence coefficient threshold value, determining that the vehicle is not positioned in the designated parking area.

As an example, the visual positioning indicator includes a first positioning indicator and a second positioning indicator, and the first positioning indicator is preset to have a weight of 40% and the second positioning indicator is preset to have a weight of 60%. The main control unit performs weighted summation to obtain the confidence coefficient that the vehicle is located in the designated parking area, namely determining the product of the confidence coefficient of the detection frame of the first positioning identifier and the weight corresponding to the first positioning identifier, and determining the product of the confidence coefficient of the detection frame of the second positioning identifier and the weight corresponding to the second positioning identifier, and summing the two products to obtain the confidence coefficient that the vehicle is located in the designated parking area.

Therefore, the main control unit can determine whether the vehicle is located in the designated parking area according to the visual positioning identification in the acquired environment image around the vehicle, and the problems that the positioning accuracy is poor, the non-motor vehicle is locked by mistake or the locking fails and the like due to the fact that the GPS signal drifts in the complicated environment area are avoided.

Because the parking direction and the parking posture of the vehicle also influence the city appearance of the city and cause the waste of parking spaces, after the main control unit determines that the vehicle is positioned in the designated parking area, whether the parking direction of the vehicle meets the direction requirement or not and/or whether the parking posture of the vehicle meets the posture requirement or not can be determined according to the obtained visual positioning identification in the environment image around the vehicle.

For example, the main control unit may determine whether the parking direction of the vehicle meets the direction requirement and determine whether the parking posture of the vehicle meets the posture requirement according to the structured information of the visual positioning identifier in the environment image. The method for determining whether the parking direction of the vehicle meets the direction requirement and determining whether the parking posture of the vehicle meets the posture requirement by the main control unit according to the structured information may refer to step 204 below, which is not repeated herein.

And step 207, if the main control unit determines that the vehicle is located in the designated parking area, locking the vehicle according to the locking instruction.

As one example, the master control unit may effect locking or unlocking of the vehicle by modifying the state of the vehicle. For example, the state of the vehicle may include a locked state and an unlocked state. The main control unit can modify the state of the vehicle from the unlocked state to the locked state according to the locking instruction, so that the vehicle is locked.

As one example, the master control unit may lock the vehicle by a locking device. For example, as shown in fig. 1, the parking management system further includes a locking device, the main control unit sends a locking instruction to the locking device if it is determined that the vehicle is located in the designated parking area, and the locking device receives the locking instruction sent by the main control unit and locks the vehicle according to the locking instruction.

For example, the locking device further includes a locking module, the locking module may be a mechanical device, and the locking module mechanically locks the vehicle if receiving a locking instruction.

As one example, if it is determined that the vehicle is not located within the designated parking area, the main control unit issues first warning information for prompting the user to adjust the parking position of the vehicle.

For example, as shown in fig. 1, the parking management system further includes an alarm module, the main control unit may send a first alarm instruction to the alarm module, and the alarm module receives the first alarm instruction sent by the main control unit and sends out first alarm information according to the first alarm instruction. The first alarm instruction is used for indicating the alarm module to send first alarm information.

In addition, before the vehicle is locked according to the lock instruction, the main control unit may further determine whether at least one of a parking direction and a parking posture of the vehicle meets a requirement. And if the vehicle is determined to be located in the designated parking area and at least one of the direction requirement and the posture requirement meets the requirement, locking the vehicle according to the locking instruction. Therefore, on the basis of determining that the vehicle is located in the designated parking area, the vehicle is locked after the parking method of the vehicle meets the direction requirement and/or the parking posture of the vehicle meets the posture requirement, parking management of non-motor vehicles can be further performed, the influence of the parking direction or the parking posture of the vehicle on city appearance of a city is reduced, and the waste of parking space resources is reduced.

For example, the main control unit may determine whether the parking direction of the vehicle meets the direction requirement according to the structured information, and may also determine whether the parking posture of the vehicle meets the posture requirement according to the structured information. And if the main control unit determines that the vehicle is positioned in the designated parking area, the parking direction meets the direction requirement and the parking posture meets the posture requirement, the main control unit executes the step of locking the vehicle according to the locking instruction.

As one example, the visual location identifier includes at least one location identifier, and the structured information includes an identifier category for each of the at least one location identifier, and a pixel location, a detection box, and a confidence level for the detection box in the environmental image. The main control unit can determine whether the parking direction of the vehicle meets the direction requirement according to the identification category of each positioning identification in the at least one positioning identification and the pixel position in the environment image.

The parking direction comprises a forward parking direction and a reverse parking direction, and the condition that the parking direction meets the direction requirement means that the vehicle is parked in the forward parking direction.

The main control unit can determine the position relation between the at least one positioning identifier according to the structured information. And if the position relation meets the first preset condition, determining that the parking direction of the vehicle meets the direction requirement. And if the position relation does not meet the first preset condition, determining that the parking direction of the vehicle does not meet the direction requirement.

For example, the main control unit determines the position relationship between at least one positioning identifier according to the identifier category of each positioning identifier in the at least one positioning identifier and the pixel position in the environment image.

As one example, the visual locator marking includes a first locator marking and a second locator marking, the positional relationship between the first locator marking and the second locator marking being such that a printed location area of the first locator marking is closer to a boundary of the designated parking area. The pixel position refers to a central pixel position of each positioning identifier in the at least one positioning identifier in the environment image, and a coordinate origin of a pixel coordinate system corresponding to the environment image is located at the upper left corner of the image. Therefore, if the main control unit determines that the pixel position of the first positioning identifier in the environment image is smaller than the pixel position of the second positioning identifier, namely the first positioning identifier is closer to the boundary of the environment image, the parking direction of the vehicle is determined to be the forward parking direction, and the parking direction of the vehicle meets the direction requirement.

As an example, the master control unit may also determine the location relationship between the at least one positioning identifier according to an identifier attribute of each of the at least one positioning identifier. For example, the main control unit determines an identifier attribute of each of the at least one location identifier through the identifier recognition model, then determines an identifier category according to the identifier attribute, and then determines a position relationship between the at least one location identifier according to the identifier category.

As an example, if the main control unit determines that the parking direction of the vehicle does not meet the direction requirement, the main control unit may send second warning information, where the second warning information is used to prompt the user to adjust the parking direction of the vehicle.

For example, as shown in fig. 1, the parking management system further includes an alarm module, the main control unit may send a second alarm instruction to the alarm module, and the alarm module receives the second alarm instruction sent by the main control unit and sends out second alarm information according to the second alarm instruction. And the second alarm instruction is used for indicating the alarm module to send second alarm information.

As one example, the visual location identifier includes at least one location identifier, and the structured information includes an identifier category for each of the at least one location identifier, and a pixel location, a detection box, and a confidence level for the detection box in the environmental image. The main control unit can determine whether the parking gesture of the vehicle meets the gesture requirement according to the detection frame of each positioning identifier in the at least one positioning identifier in the environment image and the confidence of the detection frame.

The image acquisition device does not rotate along with the rotation of the head of the vehicle, and the posture of the image acquisition device is consistent with that of the body of the vehicle, so that the width and height of the detection frame corresponding to each positioning mark in at least one positioning mark in the environment image acquired by the image acquisition device can determine the parking posture of the vehicle.

The parking posture of the vehicle refers to an angle of a vehicle body of the vehicle deviating from a preset direction. The preset direction refers to a direction set for the specification of parking of the vehicle. For example, the visual positioning identifier includes a first positioning identifier and a second positioning identifier, the first positioning identifier and the second positioning identifier are both rectangles, and the preset direction may be a direction perpendicular to the first positioning identifier and the second positioning identifier.

Wherein the main control unit can determine the parking posture of the vehicle according to the structured information. And if the parking posture meets the second preset condition, determining that the parking posture of the vehicle meets the posture requirement. And if the parking gesture does not meet the second preset condition, determining that the parking gesture of the vehicle does not meet the gesture requirement.

Wherein, the second preset condition refers to an angle threshold value that the vehicle can deviate from the preset direction. For example, the second preset condition may be 30 degrees, that is, if the parking posture of the vehicle is less than 30 degrees, it is determined that the parking posture of the vehicle meets the posture requirement.

For example, the main control unit determines a positioning identifier with the maximum confidence of the detection frame from the at least one positioning identifier as a target positioning identifier, and determines the parking posture of the vehicle according to the width and height of the detection frame of the target positioning identifier.

As one example, the parking posture of the vehicle may be determined by the following formula (1):

where θ is the parking attitude of the vehicle, h is the height of the detection frame of the target location indicator, and w is the width of the detection frame of the target location indicator.

As one example, the visual positioning markers include a first positioning marker and a second positioning marker, both of which are rectangles. The main control unit can also determine the parking posture of the vehicle according to the slope of the first positioning mark or the second positioning mark in the environment image.

As an example, if it is determined that the parking posture of the vehicle does not meet the posture requirement, the main control unit may send third warning information, where the third warning information is used to prompt the user to adjust the parking posture of the vehicle.

For example, as shown in fig. 1, the parking management system further includes an alarm module, the main control unit may send a third alarm instruction to the alarm module, and the alarm module receives the third alarm instruction sent by the main control unit and sends third alarm information according to the third alarm instruction. And the third alarm instruction is used for indicating the alarm module to send third alarm information.

It should be noted that, on the basis of determining that the vehicle is located in the designated parking area, after determining that the parking method of the vehicle meets the direction requirement and the parking posture of the vehicle meets the posture requirement, the vehicle is locked, so that the parking management of the non-motor vehicle can be further performed, the influence of the parking direction and the parking posture of the vehicle on the city appearance of the city is reduced, and the waste of parking space resources is reduced.

In addition, the above steps 205 to 207 may also be executed by a computer device outside the parking management system, where the computer device may be a terminal or a server, and the computer device stores the identification recognition model. For example, after receiving the environment image in step 204, the main control unit sends the environment image to a computer device outside the system, and the computer device executes step 205 and step 207 according to the received environment image.

If the computer device executes the step 207 and determines that the vehicle is located in the designated area, the vehicle can be locked according to the locking instruction by the following steps:

step 1) if the computer equipment determines that the vehicle is located in the designated area, generating a locking instruction and sending the locking instruction to a main control unit of the parking management system.

Wherein, the computer equipment and the main control unit can communicate through the communication device. For example, the vehicle management system includes at least a main control unit and a communication device. The computer equipment can be connected with the communication device in a wireless mode or a wired mode, and the communication device can be connected with the main control unit in a wireless mode or a wired mode.

The computer equipment can send a locking instruction to the communication device, and the communication device receives the locking instruction sent by the computer equipment and sends the locking instruction to the main control unit.

The locking instruction comprises a vehicle identifier and is used for indicating a communication device of the corresponding vehicle to receive the locking instruction.

And step 2) the main control unit receives a locking instruction sent by the computer equipment and locks the vehicle according to the locking instruction.

Therefore, the calculation amount of the main control unit in the parking management system can be reduced, and the efficiency of parking management is improved.

In the embodiment of the application, if a locking instruction of a vehicle is received, an environment image around the vehicle, which is acquired by an image acquisition device configured for the vehicle, is acquired, then the environment image is used as an input of an identification recognition model, structural information of a visual positioning identification in the environment image is determined through the identification recognition model, then whether the vehicle is located in a specified parking area is determined according to the structural information, and if the vehicle is determined to be located in the specified parking area, the vehicle is locked according to the locking instruction. Wherein the structured information is used to indicate an identification feature of the visual positioning identification. Therefore, the visual positioning identifier is preset in the designated parking area, whether the vehicle is located in the designated parking area is determined according to the visual positioning identifier in the acquired environment image around the vehicle, whether the vehicle is located in the designated parking area can be determined by combining a GPS (global positioning system), namely, parking management of non-motor vehicles by combining the GPS is avoided, and the problems that the positioning accuracy is poor, the non-motor vehicles are locked by mistake or the locking fails and the like due to the fact that GPS signals drift in a complex environment area are avoided.

In addition, whether the parking direction and the parking gesture of the vehicle meet the specifications or not can be determined according to the visual positioning identification in the environment image, so that the non-motor vehicle is further subjected to parking management, the influence of the parking direction and the parking gesture of the vehicle on city appearance and city appearance of a city is reduced, and the waste of parking space resources is reduced.

Referring to fig. 3, fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the present disclosure. As shown in fig. 3, the computer apparatus includes: a processor 301, a memory 302 and a computer program 303 stored in the memory 302 and executable on the processor 301, the steps in the parking management method in the above embodiments being implemented when the processor 301 executes the computer program 303.

The computer device may be the master control unit in the above-described embodiment of fig. 1 and the above-described embodiment of fig. 2. In a specific implementation, the computer device may be a terminal or a server, and the embodiment of the present application does not limit the type of the computer device. Those skilled in the art will appreciate that fig. 3 is merely an example of a computing device and is not intended to limit the computing device and may include more or less components than those shown, or some components in combination, or different components, such as input output devices, network access devices, etc.

The Processor 301 may be a Central Processing Unit (CPU), and the Processor 301 may also be other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or any conventional processor.

The storage 302 may be an internal storage unit of the computer device, such as a hard disk or a memory of the computer device, in some embodiments. The memory 302 may also be an external storage device of the computer device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), etc. provided on the computer device. Further, the memory 302 may also include both internal storage units of the computer device and external storage devices. The memory 302 is used to store an operating system, application programs, a Boot Loader (Boot Loader), data, and other programs. The memory 302 may also be used to temporarily store data that has been output or is to be output.

An embodiment of the present application further provides a computer device, where the computer device includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, the processor implementing the steps of any of the various method embodiments described above when executing the computer program.

The embodiments of the present application also provide a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

The embodiments of the present application provide a computer program product, which when run on a computer causes the computer to perform the steps of the above-described method embodiments.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a separate product, may be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above method embodiments may be implemented by a computer program, which may be stored in a computer readable storage medium and used by a processor to implement the steps of the above method embodiments. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include at least: any entity or apparatus capable of carrying computer program code to a photographing apparatus/terminal device, a recording medium, computer Memory, ROM (Read-Only Memory), RAM (Random Access Memory), CD-ROM (Compact Disc Read-Only Memory), magnetic tape, floppy disk, optical data storage device, etc. The computer-readable storage medium referred to herein may be a non-volatile storage medium, in other words, a non-transitory storage medium.

It should be understood that all or part of the steps for implementing the above embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. The computer instructions may be stored in the computer-readable storage medium described above.

In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus/computer device and method may be implemented in other ways. For example, the above-described apparatus/computer device embodiments are merely illustrative, and for example, a module or a unit may be divided into only one logical function, and may be implemented in other ways, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

The above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.

Claims

1. A method of parking management, the method comprising:

2. The method of claim 1, wherein the visual location marker comprises at least one location marker, the structured information comprising an identification category for each of the at least one location marker, and a detection box and a confidence level of the detection box in the environmental image;

and determining whether the vehicle is located in the designated parking area according to the identification category of each positioning identification in the at least one positioning identification and the confidence degree of the detection frame.

3. The method of claim 2, wherein said determining whether the vehicle is located within the designated parking area based on the identification category of each of the at least one location identification and the confidence level of the detection box comprises:

according to the confidence degree of the detection frame of each positioning identification in the at least one positioning identification and the weight corresponding to each positioning identification, carrying out weighted summation on the confidence degrees of the detection frames of the at least one positioning identification to obtain the confidence degree that the vehicle is located in the designated parking area:

and if the confidence level that the vehicle is located in the designated parking area is less than or equal to the confidence level threshold value, determining that the vehicle is not located in the designated parking area.

4. The method of claim 1, wherein prior to locking the vehicle in accordance with the lock instruction, the method further comprises:

and if the vehicle is determined to be located in the designated parking area, the parking direction meets the direction requirement, and the parking posture meets the posture requirement, executing the step of locking the vehicle according to the locking instruction.

5. The method of claim 4, wherein the visual location marker comprises at least one location marker, the structured information comprising an identification category for each of the at least one location marker, and pixel locations, detection boxes, and confidence levels for the detection boxes in the environmental image;

and determining whether the parking gesture of the vehicle meets the gesture requirement according to the detection frame of each positioning identifier in the at least one positioning identifier in the environment image and the confidence of the detection frame.

6. The method of claim 5, wherein determining whether the parking direction of the vehicle meets the direction requirement based on the identification category and the pixel location in the environmental image of each of the at least one location identification comprises:

7. The method of claim 5, wherein determining whether the parking pose of the vehicle meets the pose requirements based on the detection frame and the confidence level of the detection frame for each of the at least one location markers in the environmental image comprises:

determining a positioning identifier with the maximum confidence degree of the detection frame from the at least one positioning identifier as a target positioning identifier;

8. The method of any of claims 1-7, wherein the identity recognition model comprises an identity detection module and an identity classification module, the visual location identity comprises at least one location identity, and the structured information comprises an identity class for each of the at least one location identity, and confidence levels for pixel locations, detection boxes, and detection boxes in the environmental image;

the taking the environment image as an input of an identification recognition model, and determining the structural information of the visual positioning identification in the environment image through the identification recognition model comprises:

9. The method of any one of claims 1-7, wherein prior to said inputting said environmental image as an identification recognition model by which to determine structured information of a visual positioning identification in said environmental image, said method further comprises:

acquiring a sample environment image around a sample vehicle;

labeling the sample environment image to obtain sample structural information of the visual positioning marks in the sample environment image, wherein the sample structural information is used for indicating the sample identification features of each positioning mark in at least one positioning mark;

10. The method of claim 9, wherein the training a to-be-trained identification recognition model according to the sample environment image and the sample structural information to obtain the identification recognition model comprises:

and adjusting the model parameters of the identification model to be trained according to the identification errors, and taking the identification model to be trained after model parameter adjustment as the identification model.

11. The parking management system is characterized by comprising an image acquisition device and a main control unit, wherein the image acquisition device is fixedly connected with a vehicle;

12. The system of claim 11, wherein the capturing angle of the image capturing device is within a preset angle range for controlling the environment image captured by the image capturing device to include the ground environment in front of the vehicle.